JP2011170679A - Virtual computer system and resource distribution control method of the same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To change resources to be regularly allocated to virtual computers, in response to a temporary service request from an individual user. <P>SOLUTION: The virtual computer system includes: a WEB server 56 receiving a processing request from a user; a resource allocation subsystem specifying proportion of resources to be allocated to individual virtual computers 20a and the like, based on an instruction from the WEB server; and a use/allocation status display table accumulating information showing an allocation user and a resource allocation status for each virtual computer. When a processing request for changing a computer performance to a higher or lower level than a desired computer performance level registered in advance in a system is transmitted from the user to the WEB server 56, the resource allocation subsystem refers to the use/allocation status display table, extracts free resources meeting the computer performance level requested in the processing request, and changes allocation of resources including the individual virtual computer 20a and the like allocated to the user, to allocate it to the user. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention relates to a virtual machine system and a resource allocation control method thereof, and more specifically, a plurality of individual virtual machines are constructed in the virtual machine system, and each individual virtual machine is assigned to each of a plurality of individual users. The service request from each individual user is received by the individual virtual machine management subsystem using the WEB service, and the resources normally allocated to each individual virtual machine can be changed upon a temporary service request of the individual user. The present invention relates to a virtual computer system and a resource allocation control method thereof.

  Conventionally, a virtual machine system in which a plurality of virtual machines (guest VMs) can be operated on a real machine by a control program known as a virtual machine monitor program (VM monitor), and each guest VM can be operated by an operating system OS. It is known as one of the techniques for using large computers. By applying such virtual machine technology, it is possible to provide a system in which a plurality of individual virtual machines are constructed on one large virtual machine and the individual virtual machines are used by individual users.

  For example, the following Patent Document 1 (Japanese Patent Laid-Open No. 5-324361) discloses a virtual machine system in which commands executed by a guest OS on each virtual machine are handled in an integrated manner. This virtual machine system enables the input of operating system commands that run on the virtual machine to the control screen displayed by the hypervisor of the virtual machine system. The control screen displays the command and each virtual machine on which it is executed. The identifier is inputted as a pair, and the command and its execution condition can be set and changed.

  Patent Document 2 below (Japanese Patent Laid-Open No. 2008-293117) discloses a method and apparatus invention for monitoring the performance of a virtual machine. In this monitoring method and apparatus, the computer executes a virtualization program that causes each of the resources of the computer that are logically divided to operate as a first virtual computer and a second virtual computer, and the first virtual computer includes: The first OS is executed, the second virtual machine executes the second OS, and the method includes information about the resources allocated to the first virtual machine and the second virtual machine by the virtualization program, Information acquired from a virtualization program, information indicating the performance of the first virtual machine is acquired from the first OS, information indicating the performance of the second virtual machine is acquired from the second OS, and the acquired information , Information indicating the time when the information was acquired is stored in a storage device, and information indicating the time, information regarding the resource acquired at the time, and the performance are indicated. Is obtained so as to output the information.

  When a plurality of guest VMs are operated on a plurality of CPUs in such a virtual machine system, the CPU resource allocation amount (allocation) is allocated to each guest VM so that the guest VM can effectively use the CPU capacity. Ratio) and the guest VMs are run in order according to the distribution amount.

  Further, in the following Patent Document 3 (Japanese Patent Laid-Open No. 5-324361), even if the CPU allocation is uneven, a virtual computer that maintains the specified CPU resource allocation ratio and effectively uses the real CPU. A system CPU control scheme is disclosed. This control method is provided in the CPU allocation information area 16, stores the number of times the designated CPU resource allocation amount has been consumed by the operation counter for each guest VM1 to VMn, and the CPU resources of the guest VM1 to VMn on the ready queue The distribution order of the operation counters is arranged in ascending order of the number of operations of the operation counter. When the number of operations of the operation counter is the same, the CPUs of the guest VM1 to VMn are distributed in the order of the distribution ratio, and the number of operations differs. In this case, the CPU distribution of the guest VM1 to VMn is performed in the order of the small number of operations.

  In a virtual machine system, in order to prevent a decrease in processing performance as a whole system, the amount of processing given is measured in advance, and the amount of processing allocated to each virtual machine is limited so that the load is appropriate for each individual virtual machine It is necessary to take preventive measures such as controlling the load in real time so that the operating status of the virtual machine is measured in real time and the processing capacity is not reduced.

  However, in some types of operating systems (hereinafter referred to as “OS”), data indicating the operating status of the central processing unit (CPU) can be extracted, but there are inconsistencies that cannot be performed in other types of OSs. It was not possible to control as intended. Although it is technically possible to precisely measure and control the load, the control subsystem becomes enormous and requires a cost close to that of the virtual machine mechanism that provides the main functions. It is a big obstacle. As a general technique for simple load balancing, there are technical fields where, for example, the round robin method is proposed and used for cases where precise measurement is difficult, but it is not always appropriate for the virtual computer field. That's not true.

Japanese Patent Laid-Open No. 5-324361 JP 2008-293117 A Japanese Patent Laid-Open No. 5-324361

  In an information processing system in which a large number of individual virtual machines are constructed on a large virtual machine constructed in a single virtual machine and the individual virtual machines are used for arbitrary use by individual users, the overall processing When trying to change the processing capacity of an individual virtual machine without degrading the performance, there has been difficulty in commercial feasibility with the load measurement techniques and control methods proposed so far.

  Theoretically, it is possible to easily balance the load by measuring the load status and processing capacity of all individual virtual machines using known technology, but the individual virtual machine users are temporarily in high performance. Provide commercial solutions that respond to the detailed demands of saving resources by using virtual machines or temporarily moving to a low-performance virtual machine when no operation is performed for a certain period of time It has not been done.

  The work of users who use computers is diverse. Some users use advanced OSs that require large storage capacities, while others use general OSs that do not. The same application is used by the user. There are users who can process business using only general applications, and there are users who process business using specialized and advanced applications. For example, general applications can be used for operations such as creation of documents and tables, spreadsheets, and simple image processing, and high CPU processing capacity and memory are not required. On the other hand, work such as advanced image processing requires an application for that purpose, and requires high CPU processing capacity and memory.

  In addition, even if the same user is seen in units of one day, one week, or one month, it does not require a high level of CPU processing ability on a regular basis, and it uses professional and advanced applications. There are also times when processing general document creation and other tasks.

  When assigning individual virtual machines to individual users and considering the allocation of resources to be assigned to each individual virtual machine, in the case of an organizational unit in charge of a specific business within a specific company or a specific company, the processing of each individual virtual machine The ability and the resources to be allocated can be averaged based on the average work of each user, and even if so, the effective use of the resources is not significantly hindered. However, when considering a service that provides individual virtual machines to an unspecified number of individual users, the level required by individual users varies, and the system that provides standard individual virtual machines is flexible. There is a problem that it is not possible to cope with it effectively and the effective use of resources is hindered.

  Therefore, the object of the present invention is to easily change the processing capacity of the individual virtual machine according to the request of the user of the individual virtual machine in addition to the processing capacity of the individual virtual machine and the resource allocation control procedure set in advance, To provide a virtual machine system and its resource allocation control method capable of realizing rational resource allocation from the viewpoint of commercialization without increasing the overall load and without providing excessive resources. .

  That is, the present invention constructs a plurality of individual virtual machines in a virtual machine system, assigns each individual virtual machine to each of a plurality of individual users, and service requests from each individual user individually using the WEB service. It is an object of the present invention to provide a virtual machine system and a resource allocation control method thereof that can change the resources received by the virtual machine management subsystem and normally distributed to each individual virtual machine upon a temporary service request of an individual user. To do.

In order to solve the above problems, the first invention according to the present application is:
A virtual system in which a total resource of one or more real computers is logically divided, and a plurality of individual virtual computers, each of which is independent of each other, are constructed on the one or more real computers and controlled by a hypervisor. A computer system,
The virtual machine system includes a WEB server that accepts a user's processing request, a resource allocation subsystem that specifies allocation of allocated resources to the individual virtual machine based on an instruction from the WEB server, and an individual virtual machine A usage allocation status display table storing information indicating the allocation user, usage status, and resource allocation status,
Assigning to the user the individual virtual machine to which resources sufficient to satisfy a desired computer performance level previously registered in the system by the user are allocated;
When a processing request accompanied by a change to a computer performance level that exceeds or falls below the desired computer performance level registered in advance in the system is sent from the user to the WEB server, the resource allocation subsystem displays the usage allocation status display. A free resource corresponding to the computer performance level requested to be changed in the processing request is extracted with reference to the table, and the allocation of resources including the individual virtual machines allocated to the user is changed and distributed. To do.

  In one aspect of the first aspect of the invention, the resource allocation subsystem refers to a usage allocation status display table and assigns a free individual virtual computer according to a computer performance level requested to be changed in the processing request. Extracting and temporarily allocating the extracted free individual virtual machine to the user, and in one aspect of the first invention, each individual virtual machine is used by an assigned user. A system software section including an OS, each of which is assigned an internal IP address, and the resource allocation subsystem refers to a usage allocation status display table according to the computer performance level requested to be changed in the processing request. When an empty individual virtual machine is extracted, an individual virtual machine assigned to the user is assigned to the extracted individual virtual machine. The contents of the system software part of the machine are copied to the extracted individual virtual machine, and the IP address assigned to the extracted individual virtual machine is changed to the IP address assigned to the individual virtual machine assigned to the user. The extracted vacant individual virtual machines are temporarily allocated to the user by replacing them. Furthermore, in one aspect of the first invention, the resource allocation subsystem refers to a usage allocation status display table, grasps the total amount of resources that can be allocated to the virtual machine system, The amount of resources that can be allocated is calculated by calculating the difference between the two, and whether or not the request can be executed is compared with the amount of resource allocation to be added based on the requested computer performance level. Judgment is made, and notification of infeasibility is made according to the result, or resources are allocated according to the request.

The second invention according to the present application is the first invention,
The memory resource of the virtual computer system is a virtual memory in which a real memory is divided in units of a predetermined storage capacity, and a logical address is assigned to each divided unit storage area, and the virtual computer system further includes: A memory allocation table storing a virtual memory logical address allocated to each individual virtual machine and an unallocated virtual memory logical address;
When a processing request accompanied by a change to a computer performance level exceeding the desired computer performance level registered in the system is transmitted from the user to the WEB server, the resource allocation subsystem refers to the memory allocation table. Then, the logical address of the unallocated virtual memory is extracted, and the virtual memory resource is increased and temporarily allocated to the user so as to satisfy the amount of memory resource corresponding to the computer performance level requested to be changed in the processing request. Assigning and updating the memory allocation table,
When a processing request accompanied by a change to a computer performance level lower than the desired computer performance level registered in the system is transmitted from the user to the WEB server, the resource allocation subsystem refers to the memory allocation table. Among the virtual memory resources allocated to the user, the virtual memory resource is temporarily reduced to the user so as to satisfy the amount of memory resource corresponding to the computer performance level requested to be changed in the processing request. And the memory allocation table is updated with the reduced virtual memory resource unallocated.

The third invention according to the present application is
A virtual system in which a total resource of one or more real computers is logically divided, and a plurality of individual virtual computers, each of which is independent of each other, are constructed on the one or more real computers and controlled by a hypervisor. A computer system,
Among the plurality of individual virtual machines, an individual virtual machine that is temporarily allocated in response to a user's temporary processing request is configured as one or a plurality of temporarily allocated individual virtual machines,
The memory resource of the virtual computer system is a virtual memory in which a real memory is divided in units of a predetermined storage capacity, and a logical address is assigned to each divided unit storage area,
The virtual machine system includes a memory allocation table in which a logical address of a virtual memory allocated to each individual virtual machine and a logical address of an unallocated virtual memory are stored, a WEB server that receives a user processing request, A resource allocation subsystem that identifies the allocation of allocated resources to the individual virtual machines based on instructions from the WEB server, and information indicating the allocated users, usage status, and resource allocation status for each individual virtual machine A usage allocation status display table,
Assigning to the user the individual virtual machine to which resources sufficient to satisfy a desired computer performance level previously registered in the system by the user are allocated;
When a processing request accompanied by a change to a computer performance level that exceeds or falls below the desired computer performance level registered in advance in the system is sent from the user to the WEB server, the resource allocation subsystem displays the usage allocation status display. With reference to the table, free temporary allocation individual virtual machines are extracted, and referring to the memory allocation table, a change is requested in the processing request among the virtual memory resources allocated to the user. The virtual memory resource is increased or decreased so as to satisfy the amount of memory resource corresponding to the computer performance level, and is allocated to the extracted temporarily allocated individual virtual computer.

The fourth invention according to the present application is
A virtual system in which a total resource of one or more real computers is logically divided, and a plurality of individual virtual computers, each of which is independent of each other, are constructed on the one or more real computers and controlled by a hypervisor. A computer system,
Among the plurality of individual virtual machines, an individual virtual machine that is temporarily allocated to a user is configured as one or a plurality of temporarily allocated individual virtual machines,
The memory resource of the virtual computer system is a virtual memory in which a real memory is divided in units of a predetermined storage capacity, and a logical address is assigned to each divided unit storage area,
The virtual machine system includes a memory allocation table in which a logical address of a virtual memory allocated to each individual virtual machine and a logical address of an unallocated virtual memory are stored, a WEB server that receives a user processing request, A resource allocation subsystem that identifies the allocation of allocated resources to the individual virtual machines based on instructions from the WEB server, and information indicating the allocated users, usage status, and resource allocation status for each individual virtual machine A usage allocation status display table,
When the state in which the input / output operation is not performed for a certain period of time continues in the individual virtual machine allocated to the user, the resource allocation subsystem refers to the usage allocation status display table and allocates the virtual machine to the user. Is automatically released, and a temporary allocation individual virtual machine with limited resource allocation prepared in advance is automatically assigned, and the IP address assigned to the individual virtual machine allocated to the user is assigned to the user. It is characterized by temporarily assigning an individual virtual machine with limited resource allocation to the user by changing the IP address assigned to the individual virtual machine with static resource allocation.

  5th invention-8th invention concerning this application is invention of the resource allocation control method in the virtual computer system of 1st invention-4th invention.

In the virtual machine system according to the first invention of the present application, the virtual machine system specifies a WEB server that accepts a user's processing request, and allocation of allocated resources to the individual virtual machine based on an instruction from the WEB server. A resource allocation subsystem, a usage allocation status display table that stores information indicating the allocation user, usage status, and resource allocation status for each individual virtual machine,
Assigning to the user the individual virtual machine to which resources sufficient to satisfy a desired computer performance level previously registered in the system by the user are allocated;
When a processing request accompanied by a change to a computer performance level that exceeds or falls below the desired computer performance level registered in advance in the system is sent from the user to the WEB server, the resource allocation subsystem displays the usage allocation status display. With reference to the table, free resources corresponding to the computer performance level requested to be changed in the processing request are extracted, and the allocation of resources including the individual virtual machines allocated to the user is changed and distributed.

  According to this configuration, the service request from each individual user is received by the individual virtual machine management subsystem using the WEB service, and the resources normally allocated to each individual virtual machine are allocated to the temporary service request of the individual user. It can be changed easily.

  In the virtual machine system according to the second invention of the present application, in the virtual machine system according to the first invention, the memory resource of the virtual machine system divides the real memory by a unit of a predetermined storage capacity, and the divided unit A virtual memory to which a logical address is assigned for each storage area, and the virtual machine system further includes a logical address of a virtual memory allocated to each individual virtual machine and a logical address of an unallocated virtual memory. When a processing request including a memory allocation table to be stored and transmitted from the user to the WEB server with a change to a computer performance level that exceeds or falls below a desired computer performance level registered in the system in advance, the resource allocation sub The system refers to the memory allocation table and refers to the unallocated virtual memory. A logical address is extracted, and the virtual memory resources are increased or decreased temporarily to satisfy the amount of memory resources corresponding to the computer performance level requested to be changed in the processing request, and the memory allocation table is allocated to the user. Update.

  According to such a configuration, in response to a temporary processing request accompanied by a change to a computer performance level that exceeds or falls below a desired computer performance level registered in the system in advance by the user, the change is made in accordance with the requested computer performance level. Resources can be allocated to fill a certain amount of memory resources.

In the virtual computer system according to the third invention of the present application,
Among the plurality of individual virtual machines, an individual virtual machine that is temporarily assigned in response to a user's temporary processing request is configured as one or a plurality of temporarily assigned individual virtual machines,
The memory resource of the virtual computer system is a virtual memory in which a real memory is divided in units of a predetermined storage capacity, and a logical address is assigned to each divided unit storage area,
The virtual machine system includes a memory allocation table in which a logical address of a virtual memory allocated to each individual virtual machine and a logical address of an unallocated virtual memory are stored, a WEB server that receives a user processing request, A resource allocation subsystem that identifies the allocation of allocated resources to the individual virtual machines based on instructions from the WEB server, and information indicating the allocated users, usage status, and resource allocation status for each individual virtual machine A usage allocation status display table,
When a processing request accompanied by a change to a computer performance level that exceeds or falls below the desired computer performance level registered in advance in the system is sent from the user to the WEB server, the resource allocation subsystem displays the usage allocation status display. With reference to the table, free temporary allocation individual virtual machines are extracted, and referring to the memory allocation table, a change is requested in the processing request among the virtual memory resources allocated to the user. The virtual memory resources are increased / decreased and allocated to the extracted temporarily allocated individual virtual machines so as to satisfy the amount of memory resources corresponding to the computer performance level.

  According to this configuration, the service request from each individual user is received by the individual virtual machine management subsystem using the WEB service, and the resources normally allocated to each individual virtual machine are allocated to the temporary service request of the individual user. It can be changed easily. Also, in response to a temporary processing request accompanied by a change to a computer performance level that exceeds or falls below a desired computer performance level registered in the system in advance by the user, an amount of memory corresponding to the computer performance level requested to be changed Resources can be allocated to fill the resources.

In the virtual machine system according to the fourth invention of the present application,
Among the plurality of individual virtual machines, an individual virtual machine that is temporarily allocated to the user is configured as one or a plurality of temporarily allocated individual virtual machines,
The memory resource of the virtual computer system is a virtual memory in which a real memory is divided in units of a predetermined storage capacity, and a logical address is assigned to each divided unit storage area,
The virtual machine system includes a memory allocation table in which a logical address of a virtual memory allocated to each individual virtual machine and a logical address of an unallocated virtual memory are stored, a WEB server that receives a user processing request, A resource allocation subsystem that identifies the allocation of allocated resources to the individual virtual machines based on instructions from the WEB server, and information indicating the allocated users, usage status, and resource allocation status for each individual virtual machine A usage allocation status display table,
When the state in which the input / output operation is not performed for a certain period of time continues in the individual virtual machine allocated to the user, the resource allocation subsystem refers to the usage allocation status display table and allocates the virtual machine to the user. Automatic release, automatic allocation of temporary virtual machines for limited resource allocation prepared in advance, and automatic assignment of IP addresses assigned to the individual virtual machines allocated to users By changing to the IP address assigned to the resource allocation individual virtual machine, the individual virtual machine with limited resource allocation is temporarily allocated to the user.

  According to such a configuration, when the individual virtual machine originally assigned to the user continues to be in a dormant state, the surplus resources in the individual virtual machine can be used in response to a temporary processing request of another user. It can be used effectively, and the total resources prepared in the virtual machine system are not increased unnecessarily.

    According to the fifth to eighth aspects of the present application, it is possible to provide the resource allocation control method in the virtual machine systems of the first to fourth aspects, respectively.

1 is a system configuration diagram showing a configuration of a virtual computer system according to an embodiment of the present invention. It is a schematic diagram which shows the concept of the service in the virtual machine system shown in FIG. It is a block diagram which shows the structure of the computer for control in the virtual computer system shown in FIG. It is a figure for demonstrating the parameter | index showing the resource and performance of the basic virtual machine of the virtual machine system of this invention. It is a figure for demonstrating the parameter | index showing the resource and performance of the individual virtual machine allocated to a user. It is a figure which shows the structure of the usage allocation status display table which shows the usage allocation status of each individual virtual machine allocated to the user. It is a figure which shows the sequence of the process in the virtual machine system concerning the Example of this invention. It is a figure which shows the sequence of a process of the allocation subsystem when a user has the request | requirement of the individual virtual machine of a higher-performance processing capacity temporarily. It is a sequence diagram which shows the procedure of the detailed process in which an allocation subsystem redistributes a resource according to a user's process request. It is a block diagram which shows the structure of the computer for control in the virtual computer system concerning Example 2 of this invention. It is a figure for demonstrating the concept of the memory allocation in the virtual machine system concerning Example 2 of this invention. It is a figure which shows the structure of the usage allocation status display table which shows the usage allocation status of each individual virtual machine allocated to the user concerning Example 2. FIG. FIG. 13A is a diagram showing the configuration of a usage allocation status display table that indicates the usage status of each individual virtual machine allocated to a user in Embodiment 2 of the present invention, and FIG. 13A shows that there is no request to change the performance of each individual virtual computer FIG. 13B is a diagram illustrating an example of a configuration when there is a performance change request for an individual virtual machine. In Example 2 of this invention, it is a sequence diagram which shows the procedure of the detailed process in which an allocation subsystem redistributes a resource according to a user's process request.

  Hereinafter, specific examples of the present invention will be described in detail with reference to examples and drawings. However, the embodiment shown below exemplifies a virtual computer system for embodying the technical idea of the present invention, and is not intended to specify the present invention as the virtual computer system of this embodiment. However, the present invention is equally applicable to the virtual computer systems of other embodiments included in the claims.

  FIG. 1 is a system configuration diagram showing the configuration of a virtual machine system according to an embodiment of the present invention. The virtual machine system S according to the first embodiment of the present invention is shown in FIG. 1 as a basic virtual machine 10 used by a plurality of users, individual virtual machine groups 20a to 20n set as software on the virtual machine, It comprises user terminals 40a to 40n for users who use the individual virtual machines 20a to 20n to input and output, a control computer 50 for managing resource allocation according to user requests, and the like.

  The user terminals 40a to 40n transmit a request to the control computer 50 via the network 30, and use the individual virtual computers 20a to 20n assigned thereto. Based on the request received from the user terminals 40a to 40n, the control computer 50 causes the individual virtual machines 20a to 20n assigned thereto to process the request via the basic virtual computer 10. The network 30 is a so-called in-house network (intranet) when the user (user terminal) and the entity of the basic virtual machine 10 are in one organization or base, and the Internet when the user is distributed over a wide area (externally etc.). Or a wide area leased line.

  The basic virtual computer 10 performs virtualization on the hardware of a plurality of real computer groups 12 (which may be a single high-performance real computer) and the real computer group 12 (which may be a commercially available server). A hypervisor 14 (for example, configured using software called XEN developed at the University of Cambridge, UK) is mounted. In addition, a first individual virtual machine called a privileged domain that forms the basis of the individual virtual machine is configured (not shown in FIG. 1 because it is not released to the user and is an internal function). . Using the first individual virtual machine, the necessary number of second and subsequent individual virtual machines 20a to 20n for providing to the user is configured.

In configuring the second and subsequent individual virtual machines 20a to 20n, an interface corresponding to a CPU by software on the hypervisor 14 (the processor unit 24a in FIG. 1).
-24n), and an OS 26 (for example, various OSs provided by Microsoft) provided to the user is mounted on the interface. Application software (document creation software, printer driver for printing, etc.) used by the user is loaded with a compatible (compatible) version of software on the OS. Users who use the individual virtual machines 20a to 20n connect to the virtual machines assigned to the user from the user terminals 40a to 40n in FIG. 1 and use necessary applications.

  The above explanation describes the case where individual virtual machines are configured on a single real server. However, it is difficult to achieve with a single server by gathering multiple servers to form a server group. In some cases, a virtual server having a server is configured and used. That is, by using the virtualization technology, one server can be configured as a plurality of individual virtual machines, and a plurality of real servers can be combined to form a single server virtually. Using this technology, by configuring multiple individual virtual machines on a group of multiple real servers, configure the required individual computers on a high-performance server that collects multiple inexpensive servers. It is possible to construct a rational system from a commercial viewpoint.

  In a general virtual machine system, resource allocation to individual virtual machines is set so that the system side can satisfy a user's processing request. However, when considering a service that provides individual virtual machines to an unspecified number of individual users, the level required by each individual user varies, so a standard individual virtual machine is provided to users. The system can not respond flexibly and the effective use of resources is also hindered.

  FIG. 2 is a schematic diagram showing a service concept in the virtual machine system S according to the present invention shown in FIG. Here, only the user terminal 40a (user a) and the system software unit 22a of the individual virtual machine 20a assigned to the user a are illustrated. The user terminal 40a transmits a request to the control computer 50 via the network 30, and uses the individual virtual computer 20a assigned to each. Based on the request received from the user terminal 40a, the control computer 50 causes the individual virtual machines 20a assigned thereto to process the request via the basic virtual machine 10.

  An OS 26 corresponding to the user a and an individual APP (individual application) 28 used by the user a are installed in the system software unit 22a of the individual virtual machine 20a assigned to the user a. An application commonly used by a plurality of users (general) is configured to use a common APP (common application) 29 prepared on the system side. The control computer 50 has a WEB server 56, and the user terminal 40 a transmits a request to the WEB server 56 via the network 30, and uses a browser provided by the WEB server 56 to use the individual virtual machine assigned to itself. Necessary processing is performed by connecting to the computer 20a, and the progress of the processing or the processing result is stored in the individual virtual computer 20a. Management information such as the progress of processing and the amount of allocated resources is displayed on a browser provided by the WEB server 56. The configuration and function of the control computer 50 including the WEB server 56 will be described in detail later.

  When the user a executes his or her business using the computer, the user terminal 40a inputs user identification information such as a user name, PW (password), and client name registered in advance in the system and requests connection to the WEB server 56. Send. If the WEB server 56 authenticates based on the user identification information and is a legitimate user, the internal IP address of the individual virtual machine 20a assigned to the user a (inside the individual virtual machines 20a to 20n managed on the virtual machine system side) IP address), the corresponding individual virtual machine 20a is activated, the desktop screen of the individual virtual machine 20a is displayed on the display screen of the user terminal 40a using the WEB browser, and the process desired by the user a is performed. The process progress and results are also displayed on the display screen of the user terminal 20a, but the process progress and process result data are stored in the individual virtual machine 20a.

  Therefore, the user terminal 40a of the user a only needs to have a communication function using a network and a function of displaying a browser provided from the WEB server 56, and it is not necessary to install an application used by the user a. Since the OS and various application programs required for the user are upgraded on the virtual computer system side, the user does not need to perform these maintenance processes themselves. Moreover, since the countermeasure against the computer virus is constructed on the virtual machine system side, the user does not need to deal with it individually, and the security can be ensured.

  If the user terminal 40a or the like is a thin client PC compatible with a thin client system (for example, the technology disclosed in Japanese Patent Application Laid-Open No. 2005-228227), a browser is used for network connection with an individual virtual machine. Then, when displaying the result of the subsequent processing, it can be operated as a so-called thin client terminal configured by software.

  In addition, a group is formed by a plurality of users, and each user shares data created or obtained using each individual virtual machine. Access rights and access levels (display, reference, copy, rewrite) for each data If the user is set to be user-friendly, data management and data security within the group can be ensured. On the virtual machine system side, a business model using the virtual machine system is constructed by setting a price for such a service or a service provided by the individual virtual machine described below for a user or a group of users. .

  User tasks are diverse. General applications can be used for operations such as creation of documents and tables, spreadsheets, and simple image processing, and high CPU processing capacity and memory are not required. Advanced operations such as image processing require an application for that purpose, and require high CPU processing capacity and memory. Moreover, even if the same user is seen in a certain unit of time, it does not require a high level of CPU processing power constantly, and there may be times when processing business using specialized and advanced applications. For example, there are times when a general document creation process is processed.

  According to the virtual computer system S of the present invention, the user sets (registers) the processing capacity (computer performance level) of the computer that the user needs most (on average) in advance on the system side, and The resource allocation of the individual virtual machine assigned to the computer is set, and processing capacity that temporarily exceeds or falls below the processing capacity of the computer set by the user (processing capacity used on a daily basis (computer performance level)) is required. In this case, the system is requested to change the resource allocation temporarily. According to such a configuration, the user pays for the service using the individual virtual machine that receives the daily resource allocation, and the processing capacity of the computer set by the user (processing capacity for daily use) When processing capacity that is temporarily above or below is required, it is possible to receive a desired service by adding to the normal price or paying the price by reducing the normal price.

  FIG. 3 is a block diagram showing the configuration of the control computer 50. The control computer 50 is a subsystem that controls allocation of the individual virtual machines 20a to 20n in response to a user request for a higher-performance individual virtual machine or vice versa. is there. The control computer 50 includes a hypervisor IF (interface) 55, a usage allocation status display table 57, a resource allocation subsystem 58, an inter-service cooperation process 59, and the like in addition to the WEB server 56 described above. Yes.

  The hypervisor IF 55 is a communication interface for linking the basic virtual machine 10 and the control computer 50, and a user request received via the WEB server 56 realizes virtualization via the hypervisor IF. To the hypervisor 14. The processing of the hypervisor 14 in response to the user request is transmitted to the inter-service cooperation process 59 via the hypervisor IF 55.

  When the WEB server 56 receives a user request, the resource allocation subsystem 58 refers to the individual virtual machines 20a to 20n stored in the usage allocation status display table 57, determines the presence / absence of an available resource, and can be used. An instruction for selecting an individual virtual machine and providing it to the user is given to the WEB server 56 and the inter-service cooperation process 59. The configuration and details of the usage allocation status display table 57 will be described in detail later with reference to FIGS. In the inter-service cooperation process 59, the resource allocation subsystem 58 links the hypervisor 14, the WEB server 56, and the like via the hypervisor IF 55.

  Next, the concept of an individual virtual machine and its resource allocation method in the virtual machine system of the present invention will be described with reference to FIGS. FIG. 4 is an example of calculating the total availability capability index that can be allocated to the basic virtual machine 10. In the present invention, prior to resource allocation, it is necessary to grasp the total amount of computer resources that can be allocated. The background of this idea utilizes the property that the performance of the individual virtual machines 20a to 20n is determined by the amount of main memory allocated and the performance of the CPU.

  Assuming a computer with a main memory of 32GB (gigabytes) in one real computer and two CPU modules with 4 CPU cores inside (with a total of 8 CPU cores), the number of clocks is 2GHz. It is empirically known that (gigahertz) is equivalent to 30 units when converted to a laptop personal computer with a main memory of 1 GB (gigabytes) and a clock frequency of 1 GHz (gigahertz).

  Therefore, when a plurality of real computer groups 12 are used as the real computers constituting the basic virtual computer 10 in FIG. 1, the above-described conversion method is used to determine the amount of resources that can be allocated to the plurality of real computer groups 12 as a whole. Calculate with The calculation is based on the product of the main memory of the actual computer, the number of CPU cores, and the number of clocks. “÷ The number of units that can be allocated is derived using 16 arithmetic expressions. For example, in the case of 32 GB main memory, 2 CPUs, 4 cores, and 2 GHz clocks, “(32−2) × 4 × 2 × 2 ÷ 16 = 30”.

  However, in the present invention, since the purpose is to finely implement the performance of the individual virtual machines desired by the user in the distribution of the individual virtual machines 20a to 20n having different performances, the index is not managed by the number as described above. Use the management method. For example, considering the allocation of performance close to a high-performance server to individual users from the minimum resource allocation required for applications such as so-called screen savers, the above-mentioned laptop computer index is set to 4. . In the present invention, a value converted into a comprehensive index (referred to as “availability index”) based on this is used. According to this, the availability capability index of the real computer described above is expressed as “30 units × index (4) = 120”.

  The meaning of this availability capability index means that “120 ÷ 4 = 30” can be configured on the server if an individual virtual machine having a computer performance level equivalent to that of an up-to-date laptop computer is configured. However, if the lower model is a model with a processing capability of about the so-called PDA, the required availability index will be about 1, so that “120 ÷ 1 = 120 units”, which constitutes 120 units. Can be determined to be possible.

  With this calculation method, the availability capability index (an index corresponding to the computer performance level) of each real computer that can be allocated in the present invention is calculated from the physical performance values of the real computer group 12, and the availability of each real computer is calculated. By calculating the sum of the performance indexes, the total availability capability index (total performance index) of the basic virtual computer 10 can be derived. In the operation of the system of the present invention, the total availability capability index (total performance index) is dynamically switched according to each user's request, and the range not exceeding this total availability index (total performance index). And it is set as the structure which drive | works, satisfying a user's individual request | requirement. FIG. 4 exemplifies the availability capability index calculated for each real computer from number 001 to number n in this way. A value “600” obtained by summing up the availability capability indexes of the respective real computers becomes the total availability capability index of the basic virtual computer 10. Therefore, according to FIG. 4, in the virtual machine system according to the present invention, the individual virtual machines 20a to 20n can be configured within the range of the availability capability index “600”.

  FIG. 5 expresses the relationship between the availability capability index and the performance of the individual virtual machines 20a to 20n. An object of the present invention is to realize resource allocation dynamically with commercial convenience. From this point of view, performance evaluations of personal computers and portable devices actually used by users (so-called personal digital assistance = PDA) were performed. As a result, as described above, when the availability capability index required by a general-purpose laptop computer is 4, the PDA is about 1/4 of that, and the desktop personal computer corresponding to the upper model is a general-purpose laptop computer. Therefore, we decided to apply the result.

  In addition, an allocation of 0.2 as an availability capability index is also prepared in order to create a state in which driving is being performed although it is almost close to a dormant state. In addition, in order to be able to provide when a higher model (higher performance) is requested by the user, server class assignment called availability capability index 16 is also prepared. What has been described so far is the availability capability index, which is realized by changing the allocation of the real memory corresponding to each availability capability index at the time of actual system construction. FIG. 5 shows the relationship between the actual memory amounts specifically allocated to the individual virtual machines corresponding to the availability capability indexes “16” to “0.2” described above.

  FIG. 6 shows the contents of the usage allocation status display table 57 of the individual virtual machines 20a to 20n constructed as described above. This allocation table shows what availability capability index each of the individual virtual machines 20a to 20n configured in the basic virtual machine 10 at the current time has, which user is currently allocated, Whether it is in use or assigned but in an idle state, and if it is assigned for a limited time, how many hours remain and how many minutes remain This indicates which state is in the state transition that the individual virtual machines 20a to 20n can take. The state transition will be described later with reference to FIGS.

  In the present invention, in setting the number of individual virtual machines (each having a unique availability capability index), 80 individual virtual computers having an availability capability index of 4 from the business viewpoint (the basic example illustrated in FIG. 4). Of the availability capacity index 600 which is the total performance index of the virtual machines, 320 is allocated to the individual virtual machines whose availability capacity index is 4. From among the remaining availability capacity indices 280, the individual virtual computer which is the availability capacity index 8 A total of 600 availability capacity indexes are accumulated, with 20 individual capacity computers with availability capacity index 1 and 20 models with availability capacity index 0.2. Value.

  This numerical value is set from a business viewpoint, and the computer of availability capability index 8 can be replaced with 8 computers of availability capability index 1 and vice versa can be changed as a response to the user's request. it can. In the above, X, which is the total availability capability index in FIG. 6, is 600, which is a state in which the maximum allocation is made. However, among them, there are virtual machines that are not assigned to the user, and the total value Y of the availability ability indexes that are assigned and used does not exceed 600.

  In the present invention, as described above, computer performance (equivalent to a server machine, equivalent to a desktop machine, equivalent to a general-purpose laptop, etc.) desired by the user is set (registered) in advance in the system. On the system side, one of the individual virtual machines is allocated to the user based on the setting (registration) of the user, and resources (memory amount) corresponding to the availability capability index are allocated. Hereinafter, in this specification, the processing within the range of the computer performance (availability index) set by the user is referred to as a normal processing request level or a normal processing request. On the other hand, a case where the user temporarily performs processing that exceeds the normal processing request level or processing that is lower than the normal processing request level is referred to as a temporary processing request level or a temporary processing request.

  An internal IP address managed on the system side is assigned to an individual virtual machine (for example, 20a) assigned to a specific user, converted from a user name, PW, and client name to an internal IP address assigned to the user, The corresponding individual virtual machine (for example, 20a) is specified. Normally, the user requests processing from the virtual computer system within the set performance range of the computer (normal processing request level).

  When processing that exceeds or falls below the performance of the computer set by the user (temporary processing request level) is desired, this is entered when a connection request to the WEB server 56 is requested. When this request is made, the virtual machine system S refers to the use allocation status display table 57 shown in FIG. 6 and temporarily reallocates free resources. In the first embodiment of the present invention, the availability capability index corresponding to the temporary processing request level requested by the user is checked, and the availability of the individual virtual machines assigned to other users is checked. An individual virtual machine is temporarily allocated to a requesting user and used. That is, in this embodiment, when the individual virtual machine itself as a resource is in a free resource state, the free individual virtual machine is temporarily reassigned and distributed to other users.

  For example, the performance of the availability capability index “4” is registered, and the user a assigned the individual virtual machine 20a performs a temporary processing request level that requires the performance of the availability capability index “16” and is allocated to other users. When the individual virtual machine 20n having the availability capability index “16” is empty, the resource allocation subsystem 58 copies the OS 26 and application stored in the system software section of the individual virtual machine 20a to the individual virtual machine 20n ( And the internal IP address of the individual virtual machine 20n is replaced with the internal IP address of the individual virtual machine 20a. By this procedure, the user who has temporarily requested the availability capability index “16” to be changed can be temporarily allocated and changed to be used.

  Next, details of the resource allocation procedure will be described with reference to the operation sequence diagrams shown in FIGS. FIG. 7 shows an example of an individual virtual machine 20a from the group of individual virtual machines 20a to 20n provided for use by individual users in the basic virtual machine 10 having the configuration described with reference to FIGS. 3 is a sequence diagram for explaining an operation sequence of the virtual computer system S from the viewpoint of state transition.

  First, before the user a uses the individual virtual machine 20a, the user a transmits predetermined user information such as a user name, a password, and a client name to the WEB server 56, and is the user authority in the WEB server 56 as a user? It is authenticated whether or not. Since this sequence is a general process, the illustration is omitted. When the usage authority is confirmed, the communication path from the user terminal 40 is established to the individual virtual machine 20a assigned to the user a (start of use: state of sequence S100).

  The establishment of the communication path is performed by assigning an IP address dynamically assigned from the system to the individual virtual machine 20a to which the IP communication (Internet protocol communication) from the user terminal 40a is performed until the system path is established to the user terminal 40. As a part of the operation, and a normal method of establishing a communication path is performed. The user need not substantially be aware of the notified IP address, and the WEB server 56 specifies the IP address of the individual virtual machine 20a allocated to the user based on the authentication information such as the user name and PW. This is done by displaying the desktop screen of the individual virtual machine 20 corresponding to the IP address in a browser.

  When a certain period of time when the system is started up from the state of the use start sequence S100, the allocation of the individual virtual machine 20a having the predetermined performance (availability index) preset (registered in the system) by the user a is completed, and the normal resource The state transitions to the assignment state (sequence S104). Thereafter, when the individual virtual machines 20a are continuously used, the resource allocation (resource allocation) does not change particularly in the continuous operation state (sequence S110). “Continuous” means that the application program is performing a predetermined process, the user is inputting text, the input / output system of the system is operating, and the like.

  However, when such an application program has finished processing and the user does not perform key input or the like, the application program is not in use. However, while the continuous operation state (sequence S110) continues, the user activates the procedure of “cancel use” (sequence S124) or generates a “resource allocation request different from the initial allocation” (sequence The system is constructed so as to cause another state transition for each event when there is a state of (S150) or "no use for a certain period of time" (sequence S122) occurs.

  When the “cancel use” (sequence S124) procedure is activated, the communication path to the individual virtual computer 20a that has been assigned to the user until then is disconnected, and as an example shown in FIG. The usage status of the computer number 1001 is rewritten from “in use” to the usage allocation status “unassigned” and the usage status “idling” as seen in the individual virtual computer number 1003, for example, and the individual virtual computer 20a is released ( Sequence S140). As a result, the use of the individual virtual machine 20a that has been used so far is finished (sequence S142), and preparations are made available for use by other users.

  The cause of the system design for canceling the use is that the user selects to cancel the use and performs the operation for that purpose, and the result of the selection to use only for a predetermined time from the beginning is shown in FIG. The “management timer” shown in FIG. 5 may measure the remaining usable time, and a process for stopping the use may be triggered when the remaining time becomes zero.

  The occurrence of the “no use for a certain time” (sequence S122) state is a time when the user does not operate the keyboard on the terminal 40a, and the idling state is also present in the individual virtual computer 20a in use. It means a state of continuing for a certain time (set to 10 minutes in this embodiment) or longer. From the idea that the fact that CPU resources are allocated in spite of this state is a waste of resources, it is forced to make a transition to the limited resource allocation state in the dormant state (sequence S130), thereby saving the total resources. It is carried out. The procedure of this state transition is the same as when a resource allocation request different from the normal one described below is generated, and details will be described next.

  Next, a state transition when a resource allocation request different from normal occurs by the user a will be described. When a user uses a computer, a higher-performance computer may be required temporarily depending on the type of application to be used and the amount of information to be processed. For example, Adobe's Photoshop and Illustrator are frequently used as high-performance image processing software. Examples of a large amount of information to be processed are concentrated at the end of the fiscal year. Examples include accounting data for accounting software used and manuscripts for creating printing blocks.

  FIG. 8 is a sequence diagram for explaining the flow of processing when the user temporarily requests an individual virtual machine with higher performance processing power among the events of “event occurrence” (sequence S120) in FIG. It is. In order to manage the use of high-performance machines, once the high-performance processing capacity is obtained, the user's judgment may return to the original normal processing capacity. In some cases, the system can be used without any inconsistency by either method.

  When the user a needs a higher performance processing capacity, the individual virtual machine used by the user is firstly received by the WEB server 56 that receives a user connection request, a processing capacity expansion request, and the like shown in FIG. And access necessary performance requirements. From the viewpoint of ease of understanding by general users, this input is the performance that the user thinks is necessary from the “exemplary explanation of the performance of individual virtual machines distributed to the user” of FIG. 5 expressed by the WEB server 56. It is configured to be performed by a method of selecting an expression (availability ability index). For example, the performance of the individual virtual machine 20a used by the user a is "equivalent to a general-purpose laptop computer", so the availability capability index is 4, and the main memory allocation amount is 1 Gbyte. An example of the user's request (temporary processing request) this time is processing capacity “corresponding to a server machine”. In this case, the availability capability index is 16 and the main memory allocation is 4 Gbytes.

  The WEB server 56 that has received a request for using the processing capability “equivalent to the server machine” from the user converts the user's processing request in the process of the transition state (sequence S152) in FIG. 8 into the availability capability index “16”. 3 and instructing the resource allocation subsystem 58 shown in FIG. 3 that the resource required by the user a is the availability capability index “16” and searching for the virtual machine that is idling, and assigning it if available. To communicate. Next, the resource allocation subsystem 58 that has received the instruction recalculates the total value Y of the availability capability index in use in the usage allocation status display table 57 of the individual virtual machine to obtain the latest Y value ( State transition sequence S153).

  As a result, since the latest used availability index is calculated, Y + 16−4 = Y ′ is further calculated here (16 is the value of the newly allocated (requested) availability index. In the decision process 154, which is the next state transition (because there are 4 resources currently in use, the increment is being calculated), X and Y 'are compared, and if Y'> X For example, the process 156 (sequence S156) is caused to perform the process 160 (sequence S160) if Y ′ ≦ X.

  In the process 156 (sequence S156), most of the resources of the basic virtual machine 10 have already been allocated, and the availability capability index 4 that the user has used so far has been available. This means that there was no room to obtain the performance index 12. The fact that the state can be changed to the process 160 (sequence S160) means that there are more availability capability indexes that can be distributed including 12 or more.

  If the allocation is made to the process 156, the allocation is impossible, so the resource allocation subsystem 58 transmits an unallocated response to the Web server 56, and the content of the transmission is sent to the requesting user. Since the resource requested above is not currently available, the processing is stopped after displaying that it cannot meet the high performance requirement.

  Details of the process 160 (sequence S160) in FIG. 8 are shown in the sequence diagram shown in FIG. The details are shown. The resource allocation subsystem 58 that has started the process 160 (sequence S160) is requested from the usage allocation status display table 57 of the individual virtual machines 20a to 20n as shown in the process 161 (sequence S160) of FIG. An individual virtual machine assigned to another user who has the performance index 16 and whose usage status is idling is searched. That is, a search is started from the individual virtual machine number 5000 series which is a group assigned to other users as the availability capability index “16” from the individual virtual machine use allocation status display table 57 shown in FIG. Extract individual virtual machines that are idling.

  According to the example shown in FIG. 6, since the virtual machine number 5001 has the availability capability index 16 and is idling, the individual virtual machine is identified, and its internal IP address is used to display the usage allocation status of the individual virtual machine. Extracted from the table 57 and delivered to the resource allocation subsystem 58. As the process 162 (sequence S162), the resource allocation subsystem 58 creates a copy of the system software unit 22a (including the operating system 26) of the individual virtual machine 20a that requested the next allocation as a virtual machine. The virtual machine is created in the 22n portion of the individual virtual machine corresponding to the number 5001 (in this case, described as 20n in FIG. 1).

  Next, in the process 163 (sequence S163), the operation of the individual virtual machine 20n is started, and the individual virtual machine 20n is used as an IP address that replaces the connection IP address for the individual virtual machine 20a previously used as the connection destination by the terminal. The connection destination from the user terminal is stopped by performing replacement (replacement) so that the IP address of the individual virtual computer 20n extracted from the computer usage allocation status display table 57 and delivered to the resource allocation subsystem 58 can be used. It can be switched without doing.

  By changing the IP address in this way, it is possible to operate with reasonable resource allocation while temporarily changing the assigned user of the individual virtual machines having different performance indexes from the originally assigned user to another user. become. Since the individual virtual machine 20a that has been used by the user so far has been used at the stage of switching, the usage allocation status is “unallocated” on the usage allocation status display table 57 of the individual virtual machine. , The usage status is “idling”, the state description is modified, and the total availability capability index Y ′ in use is modified accordingly.

  After that, if the provided individual virtual machine 20n is steadily operated like other virtual machines, no state transition occurs. This is indicated by the state 165 (sequence (S165) in FIG. 9. However, even in this case, as described above, when the operation is not performed for a certain time (in this embodiment, 10 minutes), Is implemented so that if a transition is made to the resource allocation state 169 during suspension and any key operation is performed on the terminal (sequence S171), the steady state is returned to the steady state of sequence S164. This same procedure is configured as an automated process that is triggered by the continuation of a state where there is no operation for 10 minutes and is restored to the original state when the user operates a key on the terminal.

  Except for hibernation, the user ends the use of the high-performance individual virtual machine that has been temporarily used and returns to the use of the individual virtual machine having the standard performance index that was originally used (FIG. 9). The process 172 (sequence S172)) and the use itself are ended. When the use itself is terminated, the process proceeds to the above-described process 140 (sequence S140), and the individual virtual machine 20n that has been temporarily allocated to the user a and used is released.

  In the process 172 (sequence S172) that returns to the normal resource allocation state, the procedure is the same as that of the process 160 (sequence S160), but the resource by the evaluation of the total performance index performed in the process 154 (sequence S154). The process for determining whether assignment is possible is omitted. This is because, when switching from the use of a high-performance machine to a virtual machine with a lower performance, the necessary amount of resources always decreases, and therefore it is not necessary to determine whether or not to allocate. As a result, the processing time of the return process is shortened, and there is no inconvenience to the user. In the process 172 (sequence S172), the system software part (core image) of the individual virtual machine 20n is individually stored in the idling having the standard performance index assigned to the requesting user a (user terminal 40a). A replica is created on the virtual machine 20a, the input / output network IP addresses are changed, and the reassignment is completed.

  As described above in detail, according to the virtual machine system according to the first embodiment of the present invention, the user a who has registered the performance of the availability ability index “4” and assigned the individual virtual machine 20a can obtain the availability ability index. When the temporary processing request level that requires the performance of the index “16” is performed and the individual virtual computer 20n of the availability capability index “16” allocated to another user is empty, the resource allocation subsystem 58 The OS 26 and applications stored in the system software section of the computer 20a are copied (copied) to the individual virtual computer 20n, and the internal IP address of the individual virtual computer 20n is changed to the internal IP address of the individual virtual computer 20a. By this procedure, the user who has temporarily requested the availability capability index “16” to be changed can be temporarily allocated and changed to be used.

  In the virtual machine system according to the first embodiment, the individual virtual machine (for example, the user assigned to the user in accordance with the availability capacity index (for example, the availability capacity index “4”) set (registered) in advance in the system by the user) When a temporary processing request exceeding the availability capability index is desired instead of a normal processing request using 20a), an availability capability index satisfying the temporary processing request assigned to another user or unallocated (for example, Since the individual virtual machine having the availability capability index “16” is temporarily allocated to the requesting user, the allocation cannot be changed unless the individual virtual computer having the availability index is available. (Refer to sequence S156 in FIG. 8.) The same applies to a temporary processing request that falls below the set availability capability index.

  However, the above inconvenience does not occur if free resources that can satisfy the temporary processing request can be provided in the entire resources of the virtual computer system S. For example, in the above case where the availability capability index set (registered) in advance is “4” and the requested availability capability index is “16”, the availability capability index “4” assigned to another user is 4 If one individual virtual machine is available, the resource can be temporarily allocated to the individual virtual machine 20a of the requesting user, and this can be operated as the individual virtual machine 20a having the performance of the availability capability index “16”. .

  Therefore, in the virtual machine system according to the second embodiment, the amount of allocated memory for performance change without changing the relationship between a plurality of users existing at a certain point in time and individual virtual machines assigned to the user. In response to the user's request, the memory that is not allocated for use at that time is dynamically searched and allocated to meet the user's request. The virtual machine system according to the second embodiment is characterized in that resources are saved by eliminating individual virtual machines that are not assigned to use without using an availability capability index.

  The configuration of the virtual computer system S according to the second embodiment and the concept of service provision are the same as those in the first embodiment shown in FIGS. FIG. 10 is a block diagram illustrating the configuration of the control computer 50 in the virtual computer system S according to the second embodiment of the invention. The control computer 50 according to the second embodiment has basically the same configuration as that of the control computer 50 according to the first embodiment shown in FIG. 3 except that a memory allocation table 60 is provided. This is different from the control computer 50 of the first embodiment. The other components are the same as those in the first embodiment, and the same components as those in the control computer 50 in the first embodiment are denoted by the same reference numerals. In order to avoid duplication of explanation, explanation of these same components is omitted.

  In this embodiment, in order to perform dynamic resource allocation with commercial simplicity, the real memory of the basic virtual computer 10 (see FIG. 1) is virtualized by the hypervisor 14 that realizes virtualization. It is converted into 70, and is divided into blocks for a certain amount (256 megabytes in this embodiment) for subsequent use, and a logical address is assigned as a serial number for each unit. FIG. 11 shows the concept of such memory allocation. As shown in FIG. 11, in the second embodiment, the first 256 megabytes has a logical address 0000, the next consecutive block (256 megabytes) has a logical address of 0001, and subsequent blocks are given continuous logical addresses. ing.

  The real memory to which logical addresses are assigned in units of blocks is dynamically allocated according to the usage status of each individual virtual machine in real time according to the availability capability index given to each of the individual virtual machines 20a to 20n. . Accordingly, in the usage allocation status display table 57 in the second embodiment shown in FIG. 12, information on the reference destination of the memory allocation table 60 is added as compared with the usage allocation status display table 57 in the first embodiment shown in FIG. ing. In FIG. 12, Z is a value obtained by subtracting the total value Y of the availability capability indexes being used from the total availability capability index X.

  FIG. 13 shows the configuration of the memory allocation table 60, and FIG. 13A shows an example of a state before a change in resource allocation (resource allocation) based on a temporary processing request from a user. In this state, the logical address of the virtualized memory 70 used by each individual virtual machine 20a to 20n indicates which logical address of the virtualized memory block is allocated on the memory allocation table. Has been. At this point, each user has not made a temporary processing request that exceeds or falls below the availability capability index set (registered) in the system.

  That is, FIG. 13A shows the memory addresses of the memories (allocated memories) used by the individual virtual machines 20a to 20n at a certain time, and the logical addresses of the unallocated memories. Is also remembered. This relationship is shown in the resource allocation status (allocation status) 72 in the time section of FIG.

  Explaining the memory allocation status at this point, the individual virtual machine 1001 used by the user # 0001 is allocated 256 megabytes of memory indicated by the logical address 0000 and 256 megabytes of memory from 0001 to 0003. A total of 1 gigabyte is in use. The availability capability index of this individual virtual machine is 4. In the present invention, for ease of management, one unit of the logical address corresponds to one unit of the availability capability index.

  Next, a case will be described in which there is a temporary processing request from a certain user that requires resources that exceed or fall below the availability capability index set (registered) in the system by the user in advance. First, when the user needs higher performance, when the WEB server 56 (see FIG. 10) that accepts the user's request is requested to expand the memory amount (temporary processing request), the second embodiment Then, information on the amount of memory to be added and price information is provided to the user, and the unit of the amount of memory (increase / decrease) that needs to be added is 256 megabytes corresponding to the availability capability index. The purpose of presenting such a memory amount to the user is to measure the ease of user judgment and the ease of system operation using the availability capability index as a unit.

  As a specific example, for example, a request to increase the availability capability index “2”, which is the amount of memory already allocated to the individual virtual machine 1003 used by the user # 0003, by 4 units to obtain the availability capability index “6” Is the case. FIG. 14 is a sequence diagram showing a detailed processing procedure for reallocating resources by the allocation subsystem 58 taking this case as a specific example.

  When the WEB server 56 accepts a request from the user, it next instructs the resource allocation subsystem 58 to change the availability capability index of the individual virtual machine 1003 to “6”. When the resource allocation subsystem 58 receives an instruction from the WEB server 56, the resource allocation subsystem 58 calculates the amount of memory whose allocation should be increased in accordance with the change request. In this case, since the initial (preset) availability capability index is “2” and the availability capability index requested to be changed is “6”, the amount of memory to be allocated additionally is the availability capability index “4”. It is determined that the amount is equivalent (sequence S261).

  The resource allocation subsystem 58 then scans the “unallocated blocks” in the memory allocation table 60 illustrated in FIG. 13A for logical address numbers that are not allocated at that time, and needs to be increased. Extract logical addresses up to the number that matches the performance index. Since four units are added this time, a total of four units of logical addresses 1013, 1014, 1017, and 1018 are extracted from the unallocated list (sequence S262).

  Next, the extracted logical address list is added to the area where the logical address of the allocated memory of the individual virtual machine number 1003 in the memory allocation table 60 is recorded. As a result, the logical address list is rewritten as 1011, 1012, 1013, 1014, 1017, and 1018, although only 1011 and 1012 before the distribution (sequence S 263). As a result, preparation for the availability capability index 6 requested by the user is completed. FIG. 13B shows the state of the memory allocation table 60 at this time.

  Next, it is necessary to validate the logical address of this memory (so that the OS of the assigned individual virtual machine can be recognized), and for the validation, the OS of the individual virtual machine is restarted. . Therefore, after rewriting the list of logical addresses, the resource allocation subsystem 58 sends a “restart” instruction to the OS via the inter-service cooperation process 59, and once shuts down the OS. The system automatically waits for the OS to start up (sequence S263).

  In the second embodiment, after the “restart” instruction is sent to the OS, it takes 45 seconds to 1 minute 30 seconds until the OS starts up and becomes ready for use. There were variations. However, considering the time required to replace the computer as hardware, the performance can be changed in a very short time. In this state, the individual virtual computer that satisfies the availability capability index requested by the user can be used for the necessary time until the processing based on the temporary processing request is completed (sequence S264). That is, if there is a continuous operation on the individual virtual machines, the operation is continued in a state where the additional allocation of resources is received (sequence S265).

  Next, terminate the use of the high-performance individual virtual machine, and if the performance (availability index) of the assigned individual virtual machine needs to be returned to the performance before the change, or terminate the use of the individual virtual machine itself There may be a choice. Sequence S266 (state 266) in FIG. 14 means that this state change is being determined. As described above, when it is necessary to change the performance, the user again changes the processing capacity to the WEB server 56 (in this case, return to the original state before the change, or change the use itself). Request to change).

  In the case of returning to the original performance, as shown in the sequence S267 of FIG. 14, the reverse of the operation performed so far is performed, and the resource allocation subsystem 58 decreases the allocated memory from the WEB server 56. The instruction is received, and it is confirmed that the logical addresses of the memories currently allocated to the individual virtual machines are 1011, 1012, 1013, 1014, 1017, and 1018. Next, it is recognized that 4 blocks 1013, 1014, 1017, and 1018 need to be reduced in order for the logical address (number of blocks) meaning the request to be “2” (difference in availability capability index).

  In the present embodiment, an algorithm is adopted in which a new logical address is added from a smaller logical address number, and a new logical address is decreased from a larger number. Accordingly, a total of four units of logical addresses 1013, 1014, 1017, and 1018 that need to be reduced are moved to an unallocated area of the memory allocation table 60. The migration here means that the logical address of the memory to be migrated is deleted from the logical address of the memory registered in the individual virtual machine 1003 on the memory allocation table 60, and the logical address is added to the unallocated area. That means.

  As a result, what was 1011, 1012, 1013, 1014, 1017, 1018 before the distribution change (see FIG. 13B), the logical address list is rewritten to the original state as 1011, 1012 (FIG. 13A). reference). Thus, preparations for returning to the availability capability index 2 requested by the user are made. Next, it is necessary to validate the logical address of the memory, and it is necessary to restart the OS of the individual virtual machine for the validation. The OS other than that used by the individual virtual machine does not perform any operation.

  As described above, the resource allocation subsystem 58 rewrites the logical address list in the memory allocation table 60, and then sends a “reboot” instruction to the OS. wait. By this procedure, the individual virtual machine returns to the individual virtual machine having the availability capability index “2”, shuts down the temporarily used OS, and waits for subsequent automatic startup. This procedure returns to the individual virtual machine with the availability capability index 2 and the temporarily used memory is released.

  In addition to this, when the selection to end the operation is performed, all of the logical addresses 1011, 1012, 1013, 1014, 1017, and 1018 allocated to the individual virtual machine 1003 are rewritten to the unallocated area. Thereafter, an instruction to “shut down” is sent from among the termination options of the OS, the operation of the OS is stopped, and the use of the individual virtual machine is terminated (sequence S268).

  As described above, according to the virtual computer system according to the second embodiment described above, in the virtual computer system according to the first embodiment, a change from a user to a computer performance level that exceeds or falls below a desired computer performance level registered in the system in advance. In response to a temporary processing request, resources can be allocated so as to satisfy an amount of memory resources corresponding to the requested computer performance level.

  As described above, according to the virtual machine system according to the second embodiment described above, when a temporary processing request exceeding or lowering the processing performance (availability index) previously set (registered) by the user is assigned to the user, The resource allocation (memory amount) allocated to the individual virtual machine is temporarily increased or decreased without changing the individual virtual machine having the predetermined availability capability index. Further, in the virtual machine system according to the first embodiment described above, when a temporary processing request that exceeds or falls below the processing performance (availability index) set (registered) in the system by the user is assigned to the user, A temporary individual virtual machine having an availability capability index satisfying the temporary processing request assigned to or not assigned to another user is temporarily assigned to the requesting user without using the assigned individual virtual machine. there were.

  In addition to the first embodiment and the second embodiment, a virtual computer system using both the configuration of the first embodiment and the configuration of the second embodiment can be constructed. The virtual computer system according to the third embodiment of the present invention constructs one or a plurality of individual virtual computers that are simply waiting for an input signal, generally called a screen saver with an availability capability index of 0.2, When an individual virtual machine originally assigned to a user has not been operated for a certain period of time (10 minutes in this embodiment), the user can use the individual virtual machine with the availability capability index of the first embodiment. In addition to applying the procedure (IP address replacement and system software copy), the memory amount increase / decrease according to the availability capability index requested by the user (procedure of the second embodiment) is also applied as required.

  A virtual machine system according to the third embodiment of the present invention will be briefly described below. In this virtual computer system S, an individual virtual computer that is simply waiting for an input signal, such as an availability capability index “0.2” as in the first embodiment, generally called a screen saver, can be operated for a certain period of time. It is intended to save total resources and use them efficiently by temporarily replacing them with individual virtual machines that are not used. The basic configuration of the virtual computer system in the third embodiment is the same as the configuration shown in FIG. 1, and the control computer is the same as the configuration of the control computer according to the second embodiment shown in FIG.

  In this embodiment, in the individual virtual machine usage allocation status display table of FIG. 11, a plurality of newly allocated virtual capacity computers with the availability capability index “0.2” are installed, and individual virtual machines are managed for the purpose of management. As computer numbers, numbers 2001, 2002 to 2010 are given. These individual virtual machine groups are segments different from each other as internal IP addresses. XXX (XXX is 1 to 10) is assigned, the IP address of the individual virtual machine number 2001 corresponds to 192.168.20.1, the address of the number 2002 corresponds to 192.168.20.2, and the others correspond to each other In this way, an address is assigned in such a manner that it can be managed separately from the individual virtual machine assigned to the user.

  Therefore, these individual virtual machines for change are exclusively used to save the resources of the individual virtual machines when the individual virtual machines originally assigned to the user are not operated for a certain period of time (referred to as hibernation). It can be used for changing the assignment of an individual virtual machine assigned to a user. With this configuration, there is no empty individual virtual machine in the first embodiment, or the use of the other user overlaps with the original user's use of the individual virtual machine that is temporarily changed and assigned to another user due to the change. The inconvenience that causes a situation where the user waits is alleviated.

  In the individual virtual machine assigned to a certain user, when the hibernation state in which no operation is performed by the user for a certain period of time continues, the resource allocation subsystem (see FIG. 10) uses one of the individual virtual machines for change to be free. The user is automatically assigned temporarily by the management by the timer (sequence S122 in FIG. 7). This allocation procedure is the sequence of S166, S167, and S169 in FIG. 9 employed in the first embodiment, after duplicating the system software section and reassigning the internal IP address, and then continuously operating the individual virtual machine for change. (Sequence S130 in FIG. 7).

  When returning from the hibernation state to the original use state, the individual virtual machine for change is released and the processing is terminated (Sequence S171 in FIG. 9 in the first embodiment). Are copied to the individual virtual machine originally assigned to the user, and the IP address is changed (the procedure is the same as the sequence S162 and S163 in FIG. 9). This allows a small amount of preset resources to be temporarily allocated to the user's temporary dormant state (which can be regarded as a temporary processing request), after which the user is assigned an individual The normal processing request can be executed by returning to the virtual machine.

  The above description is a mode in which an individual virtual machine originally assigned to a user is temporarily assigned a change virtual machine to the user when there is no user operation for a certain period of time and is in a dormant state. The present invention can also be applied when there is a temporary request that exceeds or falls below a pre-registered availability capability index. In this case, if the procedure of the first embodiment for temporarily assigning the individual virtual machine for change to the corresponding user and the procedure of the second embodiment for increasing or decreasing the memory resources assigned to the individual virtual machine for change are applied in combination. Good.

  The third embodiment is implemented in order to provide a solution to the problem that the effective utilization of resources cannot be realized when the individual virtual machine originally assigned to the user in the second embodiment is in the dormant state. This is constructed by combining the techniques of Example 1 and Example 2. However, the configuration of the combined configuration of the first embodiment and the second embodiment is not limited to the realization of the dormant state, and can be applied to a wider area. The application example is shown below.

  In the first embodiment, the availability capacity index of the individual virtual machine is set to “0.2”, “1”, 4 to “16”, which is easy to manage the second embodiment (the availability capacity index is a step in which 1 is the increment / decrement unit. It is a rough stage division compared to Depending on the user's request, it may be required to change the allocation of performance more finely. To meet such a request, for example, up to the availability capability index 4 (in this case, the distribution amount of the main memory is 1 Gbyte) When the allocation method of 1 is adopted and the user requests additional performance beyond the availability capability index 4, the user's request is further increased by controlling the main memory allocation more finely in the method of the second embodiment. A solution that faithfully responds

  As another application example, when the standard equivalent to the general-purpose laptop machine shown in FIG. 5 (availability capability index 4) is used as a standard, a unique (first) selection of which model (which availability capability index) is selected The method according to the first embodiment is used, but once the method according to the first embodiment is used, the method according to the second embodiment is adopted for the adjustment (the availability capability index change range is about plus or minus 2). A composite configuration can also be implemented.

  As described above in detail, according to the virtual machine system according to the present invention, a plurality of individual virtual machines are arranged in the virtual machine system, and the processing capability according to each user's request is obtained for commercial convenience. In general, the optimal resource allocation can be achieved at a lower cost than the conventional virtual computer system.

The virtual machine system according to the present invention can be applied to a system using an intranet in which an entity of a user (user terminal) and a basic virtual machine 10 is in one organization or base, but can be applied to a wide area ( The present invention can be suitably applied as a system that provides services to distributed users (outside the company etc.).
In addition, even if there is a generational change in the OS, etc., by intensively generating and providing the service to users of occupations and business forms that are difficult to obtain assistance from information system specialists such as individual business owners Therefore, it is possible to use higher-performance functions without replacement of information systems and personal computers, and to provide ecological solutions to environmental problems such as disposal of old models that occur when replacement is performed.

S ··· Virtual computer system 10 ··· Basic virtual computer 12 ··· Real computer 14 · · · Hypervisors 20a to 20n · · · Individual virtual computers 22a to 22n · · · System software unit 24a to 24n... Processor section 26... OS
28 ... Individual application 29 ... Common application 30 ... Network 40a-40n ... User terminal 50 ... Control computer 55 ... Hypervisor IF
56 ... WEB server 57 ... usage allocation status display table 58 ... resource allocation subsystem 59 ... inter-service cooperation process 60 ... memory allocation table 70 ... virtualized memory

Claims (14)

A virtual system in which a total resource of one or more real computers is logically divided, and a plurality of individual virtual computers, each of which is independent of each other, are constructed on the one or more real computers and controlled by a hypervisor. A computer system,
The virtual machine system includes a WEB server that accepts a user's processing request, a resource allocation subsystem that specifies allocation of allocated resources to the individual virtual machine based on an instruction from the WEB server, and an individual virtual machine A usage allocation status display table storing information indicating the allocation user, usage status, and resource allocation status,
Assigning to the user the individual virtual machine to which resources sufficient to satisfy a desired computer performance level previously registered in the system by the user are allocated;
When a processing request accompanied by a change to a computer performance level that exceeds or falls below the desired computer performance level registered in advance in the system is sent from the user to the WEB server, the resource allocation subsystem displays the usage allocation status display. A free resource corresponding to the computer performance level requested to be changed in the processing request is extracted with reference to the table, and the allocation of resources including the individual virtual machines allocated to the user is changed and distributed. A virtual machine system.   The resource allocation subsystem refers to a usage allocation status display table, extracts a free individual virtual computer corresponding to a computer performance level requested to be changed in the processing request, and extracts the extracted free individual virtual computer The virtual machine system according to claim 1, wherein the virtual machine system is temporarily allocated to the user.   Each individual virtual machine has a system software unit including an OS used by an assigned user, and is assigned an internal IP address. The resource allocation subsystem refers to a usage allocation status display table, When an empty individual virtual machine corresponding to the computer performance level requested to be changed in the processing request is extracted, the content of the system software part of the individual virtual machine assigned to the user is extracted to the extracted individual virtual machine. The extracted individual virtual machine is copied to the extracted individual virtual machine, and the extracted individual individual virtual machine is replaced with the IP address assigned to the individual virtual machine assigned to the user. The virtual machine is temporarily assigned to the user. Virtual computer system.   The resource allocation subsystem refers to the usage allocation status display table, grasps the total amount of resources that can be allocated to the virtual machine system, and calculates the difference from the already allocated resource amount. Know the amount and compare it with the amount of resource allocation that should be added based on the requested computer performance level, determine whether the request can be executed, and notify the infeasibility based on the result. 4. The virtual machine system according to claim 3, wherein resources are allocated according to the request. The memory resource of the virtual computer system is a virtual memory in which a real memory is divided in units of a predetermined storage capacity, and a logical address is assigned to each divided unit storage area, and the virtual computer system further includes: A memory allocation table storing a virtual memory logical address allocated to each individual virtual machine and an unallocated virtual memory logical address;
When a processing request accompanied by a change to a computer performance level exceeding the desired computer performance level registered in the system is transmitted from the user to the WEB server, the resource allocation subsystem refers to the memory allocation table. Then, the logical address of the unallocated virtual memory is extracted, and the virtual memory resource is increased and temporarily allocated to the user so as to satisfy the amount of memory resource corresponding to the computer performance level requested to be changed in the processing request. Assigning and updating the memory allocation table,
When a processing request accompanied by a change to a computer performance level lower than the desired computer performance level registered in the system is transmitted from the user to the WEB server, the resource allocation subsystem refers to the memory allocation table. Among the virtual memory resources allocated to the user, the virtual memory resource is temporarily reduced to the user so as to satisfy the amount of memory resource corresponding to the computer performance level requested to be changed in the processing request. The virtual computer system according to claim 1, wherein the memory allocation table is updated by allocating the reduced virtual memory resource as unallocated. A virtual system in which a total resource of one or more real computers is logically divided, and a plurality of individual virtual computers, each of which is independent of each other, are constructed on the one or more real computers and controlled by a hypervisor. A computer system,
Among the plurality of individual virtual machines, an individual virtual machine that is temporarily allocated in response to a user's temporary processing request is configured as one or a plurality of temporarily allocated individual virtual machines,
The memory resource of the virtual computer system is a virtual memory in which a real memory is divided in units of a predetermined storage capacity, and a logical address is assigned to each divided unit storage area,
The virtual machine system includes a memory allocation table in which a logical address of a virtual memory allocated to each individual virtual machine and a logical address of an unallocated virtual memory are stored, a WEB server that receives a user processing request, A resource allocation subsystem that identifies the allocation of allocated resources to the individual virtual machines based on instructions from the WEB server, and information indicating the allocated users, usage status, and resource allocation status for each individual virtual machine A usage allocation status display table,
Assigning to the user the individual virtual machine to which resources sufficient to satisfy a desired computer performance level previously registered in the system by the user are allocated;
When a processing request accompanied by a change to a computer performance level that exceeds or falls below the desired computer performance level registered in advance in the system is sent from the user to the WEB server, the resource allocation subsystem displays the usage allocation status display. With reference to the table, free temporary allocation individual virtual machines are extracted, and referring to the memory allocation table, a change is requested in the processing request among the virtual memory resources allocated to the user. A virtual computer system characterized in that virtual memory resources are increased or decreased and allocated to the extracted temporarily allocated individual virtual computers so as to satisfy an amount of memory resources corresponding to a computer performance level. A virtual system in which a total resource of one or more real computers is logically divided, and a plurality of individual virtual computers, each of which is independent of each other, are constructed on the one or more real computers and controlled by a hypervisor. A computer system,
Among the plurality of individual virtual machines, an individual virtual machine that is temporarily allocated to a user is configured as one or a plurality of temporarily allocated individual virtual machines,
The memory resource of the virtual computer system is a virtual memory in which a real memory is divided in units of a predetermined storage capacity, and a logical address is assigned to each divided unit storage area,
The virtual machine system includes a memory allocation table in which a logical address of a virtual memory allocated to each individual virtual machine and a logical address of an unallocated virtual memory are stored, a WEB server that receives a user processing request, A resource allocation subsystem that identifies the allocation of allocated resources to the individual virtual machines based on instructions from the WEB server, and information indicating the allocated users, usage status, and resource allocation status for each individual virtual machine A usage allocation status display table,
When the state in which the input / output operation is not performed for a certain period of time continues in the individual virtual machine allocated to the user, the resource allocation subsystem refers to the usage allocation status display table and allocates the virtual machine to the user. Is automatically released, and a temporary allocation individual virtual machine with limited resource allocation prepared in advance is automatically assigned, and the IP address assigned to the individual virtual machine allocated to the user is assigned to the user. A virtual computer system characterized by temporarily assigning an individual virtual computer with limited resource allocation to the user by changing to an IP address assigned to the individual virtual computer with physical resource allocation. A virtual system in which a total resource of one or more real computers is logically divided, and a plurality of individual virtual computers, each of which is independent of each other, are constructed on the one or more real computers and controlled by a hypervisor. A resource allocation control method in a computer system, comprising:
The virtual machine system includes a WEB server that accepts a user's processing request, a resource allocation subsystem that specifies allocation of allocated resources to the individual virtual machine based on an instruction from the WEB server, and an individual virtual machine A usage allocation status display table storing information indicating the allocation user, usage status, and resource allocation status,
Assigning to the user the individual virtual machine to which resources sufficient to satisfy a desired computer performance level previously registered in the system by the user are allocated;
When a processing request accompanied by a change to a computer performance level that exceeds or falls below the desired computer performance level registered in advance in the system is sent from the user to the WEB server, the resource allocation subsystem displays the usage allocation status display. Referring to a table, extracting free resources according to the computer performance level requested to be changed in the processing request, and changing and allocating resources including individual virtual machines assigned to the user; A resource allocation control method in a virtual computer system, comprising:   The step of extracting free resources according to the computer performance level requested to be changed in the processing request by the resource allocation subsystem with reference to the usage allocation status display table includes the computer performance requested to be changed in the processing request. 9. The resource allocation control method in a virtual machine system system according to claim 8, wherein the process is a process of extracting a free individual virtual machine that extracts a free resource according to a level.   Each of the individual virtual machines has a system software unit including an OS used by the assigned user, is given an internal IP address, and the resource allocation subsystem refers to the usage allocation status display table, When an empty individual virtual machine corresponding to the computer performance level requested to be changed in the processing request is extracted, the content of the system software part of the individual virtual machine assigned to the user is extracted to the extracted individual virtual machine. Copying to the extracted individual virtual machine, and replacing the extracted empty space by replacing the IP address assigned to the extracted individual virtual machine with the IP address assigned to the individual virtual machine assigned to the user. Temporarily assigning an individual virtual machine to the user. Resource allocation control method in a virtual computer system system according to claim 8,.   The resource allocation subsystem refers to the usage allocation status display table, grasps the total amount of resources that can be allocated to the virtual machine system, and calculates the difference from the already allocated resource amount. Determine if the request can be executed by comparing the step to know the amount with the amount of resource allocation to be added based on the requested computer performance level, and execute the result The resource allocation method in the virtual machine system according to claim 10, further comprising a step of notifying the impossibility or allocating resources according to a request. The memory resource of the virtual computer system is a virtual memory in which a real memory is divided in units of a predetermined storage capacity, and a logical address is assigned to each divided unit storage area, and the virtual computer system further includes: A memory allocation table storing a virtual memory logical address allocated to each individual virtual machine and an unallocated virtual memory logical address;
When a processing request accompanied by a change to a computer performance level exceeding the desired computer performance level registered in the system is transmitted from the user to the WEB server, the resource allocation subsystem refers to the memory allocation table. Then, the logical address of the unallocated virtual memory is extracted, and the virtual memory resource is increased and temporarily allocated to the user so as to satisfy the amount of memory resource corresponding to the computer performance level requested to be changed in the processing request. Allocating and updating the memory allocation table;
When a processing request accompanied by a change to a computer performance level lower than the desired computer performance level registered in the system is transmitted from the user to the WEB server, the resource allocation subsystem refers to the memory allocation table. The virtual memory resources are reduced to satisfy the amount of memory resources corresponding to the computer performance level requested to be changed in the processing request among the virtual memory resources allocated to the user. 9. The resource allocation control method in a virtual machine system according to claim 8, further comprising a step of updating the memory allocation table with no resources allocated. A virtual system in which a total resource of one or more real computers is logically divided, and a plurality of individual virtual computers, each of which is independent of each other, are constructed on the one or more real computers and controlled by a hypervisor. A resource allocation control method in a computer system, comprising:
Among the plurality of individual virtual machines, an individual virtual machine that is temporarily allocated in response to a user's temporary processing request is configured as one or a plurality of temporarily allocated individual virtual machines,
The memory resource of the virtual computer system is a virtual memory in which a real memory is divided in units of a predetermined storage capacity, and a logical address is assigned to each divided unit storage area,
The virtual machine system includes a memory allocation table in which a logical address of a virtual memory allocated to each individual virtual machine and a logical address of an unallocated virtual memory are stored, a WEB server that receives a user processing request, A resource allocation subsystem that identifies the allocation of allocated resources to the individual virtual machines based on instructions from the WEB server, and information indicating the allocated users, usage status, and resource allocation status for each individual virtual machine A usage allocation status display table,
Assigning to the user the individual virtual machine to which resources sufficient to satisfy a desired computer performance level previously registered in the system by the user are allocated;
When a processing request accompanied by a change to a computer performance level that exceeds or falls below the desired computer performance level registered in advance in the system is sent from the user to the WEB server, the resource allocation subsystem displays the usage allocation status display. A step of extracting a temporary temporarily allocated individual virtual machine with reference to the table and a request for a change in the processing request among virtual memory resources allocated to the user with reference to the memory allocation table. And increasing and decreasing the virtual memory resources so as to satisfy the amount of memory resources according to the performance level of the computer, and allocating them to the extracted temporarily allocated individual virtual computers. Distribution control method. A virtual system in which a total resource of one or more real computers is logically divided, and a plurality of individual virtual computers, each of which is independent of each other, are constructed on the one or more real computers and controlled by a hypervisor. A resource allocation method in a computer system,
Among the plurality of individual virtual machines, an individual virtual machine that is temporarily allocated to a user is configured as one or a plurality of temporarily allocated individual virtual machines,
The memory resource of the virtual computer system is a virtual memory in which a real memory is divided in units of a predetermined storage capacity, and a logical address is assigned to each divided unit storage area,
The virtual machine system includes a memory allocation table in which a logical address of a virtual memory allocated to each individual virtual machine and a logical address of an unallocated virtual memory are stored, a WEB server that receives a user processing request, A resource allocation subsystem that identifies the allocation of allocated resources to the individual virtual machines based on instructions from the WEB server, and information indicating the allocated users, usage status, and resource allocation status for each individual virtual machine A usage allocation status display table,
When a state in which input / output operations are not performed for a certain period of time continues in the individual virtual machine allocated to the user, the resource allocation subsystem refers to the usage allocation status display table and allocates the virtual machine to the user. Automatically releasing a temporary allocation individual virtual machine for limited resource allocation prepared in advance, and an IP address assigned to the individual virtual machine allocated to the user to the user A step of temporarily allocating an individual virtual machine with limited resource allocation to the user by changing to an IP address assigned to the allocated virtual machine with limited resource allocation. Resource allocation method.

Images